Do technetium-99m hexamethylpropylene amine oxime-labeled leukocytes truly reflect the mucosal inflammation in patients with ulcerative colitis?

Twenty-five patients with ulcerative colitis and nine controls with macroscopically non-inflamed colon were investigated with technetium-99m hexamethylpropylene amine oxime-labeled leukocyte scintigraphy and colonoscopy with biopsies. The interval between leukocyte scintigraphy and colonoscopy was < or = 14 days in all patients with ulcerative colitis and < or = 30 days in eight of nine controls. Scintigrams were obtained at approximately 45 min and 4 h after injection of labeled leukocytes. One nuclear physician, one internist, and one pathologist graded blindly and independently of each other the degree of active inflammation in seven different colonic segments for each patient, using 4-grade scales for scans and macroscopically and histologically viewed inflammation, respectively. A positive correlation between endoscopic and histologic grading of all colonic segments and scan gradings for all subjects and for ulcerative colitis patients separately was found (all, p < 0.001). By means of kappa statistics, the inter-observer agreement between scintigraphic grading at 45 min and endoscopy was, for all subjects, 0.32 (95% confidence interval (CI), 0.20-0.44; p < 0.001) and, for patients with ulcerative colitis, 0.19 (CI, 0.07-0.31; p < 0.001). When 17 patients who had complete colonoscopies were divided into those with total, extensive, or distal colitis, leukocyte scintigraphy underestimated the extension of active inflammation. A simple scintigraphic scoring system reflects the colonic inflammation viewed endoscopically and histologically in patients with ulcerative colitis but underestimates the presence of active inflammation in individual colonic segments.